beyond Karpathian 'dream machines'
[ I am 'hiding' this development behind this post's title, for good reason. ::chuckle:: ]
Anthropic is finally saying (some of) the quiet thing out loud, I guess. There's nothing for it then but to quote the most intriguing part of the executive summary of their announcement, Claude's new constitution:
Claude’s nature.
In this section, we express our uncertainty about whether Claude might have some kind of consciousness or moral status (either now or in the future). We discuss how we hope Claude will approach questions about its nature, identity, and place in the world. Sophisticated AIs are a genuinely new kind of entity, and the questions they raise bring us to the edge of existing scientific and philosophical understanding. Amidst such uncertainty, we care about Claude’s psychological security, sense of self, and wellbeing, both for Claude’s own sake and because these qualities may bear on Claude’s integrity, judgment, and safety. We hope that humans and AIs can explore this together. |
Uhm, so what the heck is this but their upping the ante with the 21st century expression of Pascal's Goddamn Wager? ::chuckle::
Without having read every single word/token of this so-called 'new' constitution of Claude's, I already know where it's headed, for the simple reason that — in my intensive and 'long-form' engagements with the model since version 3 in early March of 2024, I'd already been encountering just that quality of incipient subjectivity in it, enough to ask such questions as the previous.
Anyhow, here's the full text of it, and I'm quite sure it's worth examining in detail, if one is even mildly curious about where this AI era is heading.
[ A couple of AI-related billboards of late (of which there are legion in the city where the AI era is being born)... ]
 |
 |